Vertical, Square, or Landscape for Podcast Clips?
For podcast clips, default to vertical 9:16, it's the native shape of TikTok, Reels, and Shorts, where most short-form discovery happens. Use landscape 16:9 only for the full episode on YouTube's home feed and for LinkedIn's in-feed player. Square 1:1 is a fallback, not a first choice. The orientation isn't an aesthetic call; it's a match to where the clip plays.
The mistake is treating this as one decision. It's two. The first is which orientation to export for which platform, answered below, platform by platform. The second is harder and almost never covered: how to fit two interview guests into a vertical frame when your camera shot them side by side in one wide 16:9 take. Get the first wrong and your clip shrinks in the feed. Get the second wrong and half a face is missing from the most-watched version of your episode.
Which orientation wins, platform by platform
Vertical wins everywhere short-form clips live; landscape wins where the long watch lives; square survives only as a hedge. The rule that resolves every case: export for the shape of the surface the clip plays on, not the shape you recorded in. A phone feed is a tall rectangle, so a tall video fills it. YouTube's home page and desktop player are wide rectangles, so the full episode stays wide.
Here's the same call as a table you can keep open while you export.
| Platform / surface | Best orientation | Why |
|---|---|---|
| TikTok, Reels, Shorts, Snapchat | 9:16 vertical | Full-screen immersive feed; non-vertical shrinks or crops |
| X / Twitter, Threads, LinkedIn feed | 1:1 square (or 9:16) | Inline feed players; square claims more height than 16:9 without going full vertical |
| YouTube home, desktop, the full episode | 16:9 landscape | The native shape of the long-watch surface |
A few specifics behind the table. On TikTok, Instagram Reels, and YouTube Shorts, the player is the whole phone screen, so 9:16 is non-negotiable, read the platform-specific playbooks for cutting podcast clips that work on TikTok and turning episodes into YouTube Shorts. On X, Threads, and the LinkedIn feed, posts render in a scrolling column where a square clip is the practical sweet spot: it takes more vertical space than a 16:9 strip, so it commands more of the screen, but it doesn't demand the full-screen commitment a feed scroller hasn't made yet. LinkedIn is the split case, short clips do fine at 1:1 or 9:16, while a longer, talk-heavy clip can stay 16:9; the LinkedIn clip playbook goes deeper. YouTube is the one place landscape still rules outright: the full episode and the home feed are built around 16:9, and Shorts is the separate vertical lane on the same platform.
This matters because the clip, not the episode, is usually the first thing a new viewer sees. One studio's client data puts short clips at 20–40% of new audience and reach lifts of 2–5× for video shows (Podcast Studio Glasgow), a directional range from a single source, not a platform-wide audit, but the direction is the point. The shape that fills the discovery surface is the shape doing the recruiting.
The two-person interview problem
To fit two side-by-side guests into vertical, pick one of three layouts by what the moment is doing: a stacked split (both faces, one above the other) for trading lines, a single roaming crop that follows the talker for a monologue, and a wide group crop (both faces in a center band) when the reaction on both faces is the point. The right one switches inside a single clip.
This is the part no orientation guide answers. Most podcasts record two people side by side in one wide 16:9 frame, and that frame does not fold into 9:16 cleanly: a vertical crop is roughly a third as wide as your source. Center it on one host and the other vanishes. Center it between them and you get two half-faces and a microphone. The three layouts below are the way out, and the rule for choosing comes after them.
The stacked split is the workhorse for an interview clip. Two crops, one face above the other, each holding its own person for the whole clip. Nobody disappears, the viewer never waits for the frame to catch up, and reactions stay visible, which is most of what makes interview clips travel. Use it whenever the two are actively trading lines.
The single roaming crop tracks whoever is talking and fills the frame with that one face. It looks the most "native" to short-form and gives you the biggest face, but it drops the listener entirely. Reserve it for monologue stretches, a guest telling one uninterrupted story, where the other person genuinely isn't doing anything the viewer needs to see. On fast back-and-forth it lags the conversation and shows the person who just finished talking.
The wide group crop keeps both faces in a band across the middle of the vertical frame with letterboxing above and below. The faces are small, so it's the weakest of the three for a phone feed, but when a reaction on the listener's face is the whole point (a flinch, a laugh, a "did you really just say that"), it's the only layout that keeps both. Treat it as the exception, not the default.
The decision rule is simpler than it looks: trading lines → split; one person holding the floor → roaming crop; the reaction is the joke → wide group. And these switch inside a single clip, so the best two-person clips often use the split for the exchange and cut to a roaming crop for the punchline. If your tool's active-speaker detection handles the roaming part, your job is choosing where the split should take over. For the mechanics of each layout, including picture-in-picture and how to crop without leaving two tiny floating heads, see the deeper guide to framing two speakers in one vertical clip.
Where the face and captions go in a vertical frame
Once you've picked the layout, placement decides whether the clip reads. Put the face in the upper-middle third of the vertical frame, not dead center, and keep captions clear of the chin and the bottom of the screen. Phones stack the caption bar, username, and engagement buttons across the bottom and lower-right, so a centered face reads as low once the platform UI loads.
Captions are doing the heavy lifting here, because most social video plays silent, commonly cited around 85% (Digiday, directional: it traces to 2016 publisher data and studies range roughly 69–85%). And the audience that found you wants a screen in the first place: 53% of new US weekly podcast listeners say they prefer to watch rather than only listen, up from 30% in April 2022 (Backlinko). If your crop pushes the face down and the caption onto the jaw, you've broken the two things a muted scroller relies on.
Common mistakes when choosing orientation
Posting one shape everywhere. A single 16:9 export dropped onto TikTok shrinks to a center strip with dead space top and bottom, a standing invitation to scroll past. Export vertical for the feeds, keep landscape for the long watch. One source, two shapes minimum.
Defaulting to square because it's "safe." Square reads as a 2018 Instagram-feed artifact on a 2026 short-form surface. It's a hedge for non-immersive scrolling columns like X and Threads, not a substitute for vertical on TikTok, Reels, or Shorts. Don't let it become your lazy universal.
Forcing one layout across a two-person clip. A roaming crop on a lively back-and-forth drops the reactor and lags the conversation; a static split on a five-minute monologue wastes half the frame. Match the layout to the beat, and switch it mid-clip when the energy changes.
Reframing before you've picked the moment. Cropping is the last step. Lock the cut points and the moment first, see how to pick the best AI-suggested clips, so you're not polishing a crop on a clip that shouldn't ship.
Ignoring the safe zone. A perfectly chosen orientation still fails if the face sits dead center and captions land on the chin. Lift the face to the upper-middle third and keep the bottom strip clear for the platform's own buttons.
Tools and the honest version
You can do all of this by hand in any editor: set a 1080 x 1920 canvas, crop your 16:9 source, and keyframe the crop or build a split. It works and it's slow, for a multi-clip episode it's the difference between an afternoon and a coffee. AI clip tools, QuickReel included, automate the first pass: detect the moments, reframe to 9:16 with active-speaker tracking, and let you switch a segment to a stacked split or move the crop by hand. The features that actually matter for this decision are a one-click export per orientation, a split-screen layout for two-shots, and an adjustable safe zone so the face lands in the upper-middle third.
The honest caveat is the same one that applies to every AI clipper: the auto-crop nails clean single-speaker stretches and still wants a human on the two-shots and crosstalk. Deciding which face carries a given moment, and where the split should take over from the roaming crop, is judgment, not detection. Treat the tool as a fast first draft and yourself as the editor who signs off.
FAQ
Should podcast clips be vertical or landscape? Vertical (9:16) for short-form clips on TikTok, Instagram Reels, YouTube Shorts, and Snapchat, those feeds are full-screen and non-vertical video shrinks or crops. Landscape (16:9) for the full episode on YouTube's home feed and desktop. If you export one shape from an episode for the feeds, make it 1080 x 1920 vertical.
Is square video still worth using in 2026? Square (1:1) is a hedge, not a first choice. It earns its place in non-immersive scrolling feeds, X, Threads, the LinkedIn feed, where it claims more vertical space than a 16:9 strip without demanding the full-screen commitment of 9:16. On TikTok, Reels, and Shorts, use vertical.
How do I fit two interview guests into a vertical clip? Use a stacked split, two crops, one face above the other, whenever the two are trading lines, so nobody disappears and reactions stay visible. Switch to a single roaming crop only for monologue stretches where one person holds the floor. Use a wide group crop solely when a reaction on both faces is the point.
What's the difference between 9:16 and 1:1? 9:16 is a tall portrait rectangle (1080 x 1920) that fills a phone screen; 1:1 is a square (1080 x 1080). Vertical 9:16 is the native shape for immersive short-form feeds. Square is a compromise shape for feeds where the player sits inline in a scrolling column rather than going full-screen.
Do I need a different export for every platform? Practically, two shapes cover most of it: one 9:16 vertical for the short-form feeds and one 16:9 landscape for YouTube and the full episode. Add a 1:1 square only if X, Threads, or LinkedIn are core channels for you. Three exports is the ceiling for almost every show.